Mapping the bacterial cell architecture into the chromosome

A genome is not a simple collection of genes. We propose here that it can b e viewed as being organized as a 'celluloculus' similar to the homunculus o f preformists, but pertaining to the category of programmes (or algorithms) rather than to that of architectures or structures: a significant correlat ion exists between the distribution of genes along the chromosome and the p hysical architecture of the cell. We review here data supporting this obser vation, stressing physical constraints operating on the cell's architecture and dynamics, and their consequences in ter ms of gene and genome structur e. If such a correlation exists, it derives from some selection pressure: s imple and general physical principles acting at the level of the cell struc ture are discussed. As a first case in point we see the piling up of planar modules as a stable, entropy-driven, architectural principle that could be at the root of the coupling between the architecture of the cell and the l ocation of genes at specific places in the chromosome. Mie propose that the specific organization of certain genes whose products have a general tende ncy to form easily planar modules is a general motor for architectural orga nization in the bacterial cell. A second mechanism, operating at the transc ription level, is described that could account for the efficient building u p of complex structures. As an organizing principle we suggest that explora tion by biological polymers of the vast space of possible conformation stat es is constrained by anchoring points. In particular, we suggest that trans cription does not always allow the 5'-end of the transcript to go free and explore the many conformations available, but that, in many cases, it remai ns linked to the transcribing RNA polymerase complex in such a way that loo ps of RNA, rather than threads with a free end, explore the surrounding med ium. In bacteria, extension of the loops throughout the cytoplasm would the refore be mediated by the de novo synthesis of ribosomes in growing cells. Termination of transcription and mRNA turnover would accordingly be expecte d to be controlled by sequence features at both the 3'- and 5'-ends of the molecule. These concepts are discussed taking into account in vitro analysi s of genome sequences and experimental data about cell compartmentalization , mRNA folding and turnover, as well as known structural features of protei n and membrane complexes.